[torqueusers] job runs only in 1 node

Prakash Velayutham velayups at email.uc.edu
Fri Mar 10 08:16:56 MST 2006


biolabadmin at ibt.lt wrote:
> Hello,
>
> I have some problems with torque and mpich 2 maybe you could help me
>
> i want to use torque on cluster with 15 work nodes and 1 master node.
>
> i instaled Torque with maui sheduler
>
> pbsnodes -l shows all active nodes, when i try to execute job
> qsub -l nodes=4 ./try.sh , pbsnodes -l says what these 4 nodes are working
> but
> job executes only in one last node.
>
> if i use with mpich   and torque
> qsub -l nodes=4 and mpiexec -n 4 /try.sh, the i got my asked results.
> How to make torque add jobs on all asked nodes? thx
>
> or maybe you know good howto explaining all these problems?
>
> Rytis
I think you are confusing between a parallel job and multiple serial 
jobs executing in several nodes simultaneously. If you want to execute 
the same serial job on several nodes simultaneously, but through PBS, 
then mpiexec is a good way to go (as you have already figured out). If 
you do not necessarily need PBS / Torque, then look at c3 
(http://www.csm.ornl.gov/torc/C3/) from ORNL.

Prakash


More information about the torqueusers mailing list